Is it good that blitz3d directx10?(pls feedback)

Hi all,

Is it a good idea to have blitz3d supporting directx 10?

It is said there will no major change in future version of blitz3d and recommented to move to BlitzMax. End Result: Blitz3D will appear in museum (because there will be no more support to directx7 and no further development of Blitz3D soon. DxxxBasic should go to museum first).

There is no denying fact that BlitzMAX is GREAT product. But for a beginner with no or less experience, it seems that Blitz3D is much easier to pick up.

To Sir Mark: is it a good idea to have 2 product lines: Blitz3D for beginner (child or less experience), BlitzMAX for advanced users (strong oop experience and demand high speed).

(1)Blitz3D - pc support directx 10
(2)BlitzMAX - pc support directx10, opengl, mac support, oop support

(IMO, I bought Blitz3D rather than DxxxBasic because of its simpicity and stability. If it were difficult to learn, I might hesitate.)

Long life Blitz3D and BlitzMAX!!!
Thanks in advance.
Sammy
:)

No.

Blitz3d was the successor to BlitzBasic (PC version). BlitzBasic 2d was the successor to all those other Blitz Basics.
What is wrong with just moving on? If support for a whole other engine was added to B3d, it would become a bloated piece of garbage. Thus, it is revised. What is this revision? BlitzMax.
BlitzMax is the successor to Blitz3d, just as Blitz3d is the successor to Blitz2d.
Maybe BRL will release an easy BlitzMax, maybe they won't. We'll just have to wait and see.


Also, my 2 cents on using DX7: DX7 can still be made to look really excellent in many cases. On top of that, since it's old technology, it could run really well on all sorts of platforms. (Or, in some cases, not at all...). The developer can make something sort of like Darwinia without the pixel shaders; An excellent game that uses sprites and vertex colours.
A problem that I have noticed with lots of games is their lack of quantity. Instead, they go for all the latest in graphics. This means that you will be fighting the same number of bad guys as before, because the new computer has the same ammount of trouble running the new stuff as the old one had with old stuff.

I think that B3D with DX& is a good base platform, and only needs a couple of tweaks here and there to make it as good as one could hope for for a DX7 engine.

Keeping it at DX7 means you can still support older computers, and have good compatibility across the board. Blitzmax is in development to address the issue of new technology and portability. Perhaps it will eventually be DX10. The 3D module doesn't look like it will be around for at least another year, plenty of time in which to worry about which API it supports.

I only with it was coming out sooner as currently I find myself needing to use other solutions in its stead in order to get where I want. I still like to follow whats going on in blitzworld and look forward to seeing what Bmax eventually offers.

IMO Blitz3D is good as it is, having only with few (forever) missing (essential) features. Making it DX10 or even DX8 would be a really big job... and there's very few Blitz3D games out that use even full potential of the DX7 to the fullest. To my knowledge Windows Vista will run earlier games (earlier DX) too, although that might be through emulation layer... and if you have a machine that's capable of using Vista, the emulation speed won't be a problem for DX7 games and apps.

FlameDuck: May I ask what is your point of view?

Mr. Picklesworth: BlitzPlus and Blitz3D share the same syntax, so there is no pain to upgrade. However, this is not true for situation of Blitz3D and BlitzMAX.

Assuming you were a new potential customer, without game programming experience and want to write some simple game for entertainment (2 choices):

(1) XXXBasic with easy to learn commands, supporting DX9/10, large community, update frequently?
(2) BlitzMAX with OOP (need more time to pick up), 2d-module, large community, update frequently?

Which product will you buy at first glance?


If your answer is 1, how about the following:
(1) Blitz3D with easy to learn commands, supporting only DX7/8/9/10, large community, updates recently
(2) XXXBasic with easy to learn commands, supporting DX9/10, large community, update frequently?
(3) BlitzMAX with OOP (need more time to pick up), 2d-module, large community, update frequently?

After they get used to Blitz3D, it is logical to upgrade to BlitzMAX (same family member)

Thanks
Sammy
:)

Okay, you asked for it...

As for the whole DirectX thing, Mustang pretty much nailed it. Blitz3D is just fine technologically, the way it is.

Also I think your view on BlitzMAX being "more difficult" is somewhat flawed. First of all, the reason Blitz3D is as simple as you percieve it to be, is because it has limited functionality. BlitzMAX has less limits, at the cost of extra complexity. This however does not make it a more difficult language, because it is also a more abstract language.

Why do people think Blitz3D's entity system is so much better than other products they have tried? Because an entity is a abstract concept, compared to a vertexbuffer and transformation matrices. It makes it easier to understand things when you can relate to them - and it's a bonus if you don't have to understand the underlying mathematical models either, just to use the tools. That's what abstraction is good for.

The same holds true for BlitzMAX's language over Blitz3D's language. The only reason you're having difficulties understanding the object oriented paradigm, is because you've already learnt the structured programming one. For people completly new to programming, there is no difference in learning either paradigm, except that the OO one allows us to solve more complex problems in a shorter timeframe.

It doesn't help much that there aren't any really good books on the subject in general or on BlitzMAX in particular.

Also it is my oppinion that BRLs insistance on making BlitzMAX accessable to structured programming was a huge mistake on their part. Forcing the OOP on everyone would ultimately have been the right way to do things.

If I had no game programming experience I would buy the one with the prettier screensshots. Eventually that will be BlitzMAX, because it is more empowering.

It's interesting that COD2 offers either DX7 rendering as an alternative for older systems...

"This however does not make it a more difficult language, because it is also a more abstract language."
Conversely this does not make Bmax a 'better' language either, just because it's OOPS.

As a friend of mine used to say: 'I can program in Pascal in any language.'

I'm getting the feeling here that Bmax users are saying that Bmax is better because it's more complex - flawed logic, guys. Carry on in that direction to C++ if that's how you feel.
Better you should go back to reading the Blitz3D forums and showcase and see what's still being done with Blitz3D - some exceptional stuff.
And err... Bmax still has no 3D module, right?

So what I hear you saying is that Blitz3D isn't a better language because it has the Entity system, and that it wouldn't make a difference if you had to manually juggle vertexbuffers and DX surfaces, and it would be just as accessible without it?

You can drive a car with your feet if you want to, that doesn't mean it's a good idea.

It's not because it's more complex. It's because it's more abstract. Thus it lets you solve more complex problems with less effort.

Are you intentionally misreading me? C++ is a much less abstract language than pretty much any other modern programming language you can mention. Just because something has object oriented functionality, doesn't mean it also has a high degree of abstraction. Object orientation is not a silver bullet, but it helps.

That's not the point. The exceptional stuff takes more work, knowledge, understanding and resources in a procedural language than an object oriented one. Sure, you might not care, but not everyone is as l337 as you.

So what?

Flameduck-
I think Mark was bang on in making Max accessible without using OOP. I for one didn't want to go down the OOP route but warmed to it after I bought Max. Surely there are other people like me (pre-empts the one line retort 'no there aren't')

John-
Max isn't better because its more complex- there is some extra complexity that in fact makes some tasks a lot simpler with Max, but the beauty of the language is that you really don't need to touch that layer if you want to use it like previous Blitz's. I hate C++ - I find Max so much easier and that's why I'm using it. No one will disagree with you that excellent stuff has and continues to be made with B3D.

Asking for DX10 in B3D is like asking for DX7 in Qbasic. :P

I want Max3D to come out before dx10, not b3D be upgraded.
B3D might get an update if there are problems with it on Vista, it has been said around here. Other than that, the occasional bug fix when Mark gets bored.

Max is an elegant language. People get frustrated at lack of 3D (Including myself) but its not a missing feature on behalf of the language, it's a missing skill on behalf of the programmer. I am not skilled enough to create a 3D engine, thats the bottom line.

The way max works gives it the edge- you can do so much more with it than Blitz3D, 3D at present requires a bit more skill, but access to external libs is less of a pain, and works better, than in B3D enabling way more functionality in Max.

B3D not having DirectX 10 support is a shame in some ways as i'd like to be doing stuff with shaders etc etc, but there's no point just making Blitz use DX10 without getting the new features of DX10 into Blitz3D.

To be honest, the future of indi programming lies away from Blitz as far as I am concerned, i've tried BMax and released my first program and i've done some tinkerings in it. I have found BlitzMax does not suite my needs.

It is a lower level language and although I got on well enough with it to knock out my first simple game in a few days, the fact is in B3D I knocked out my first simple (near identical) game in under a day. BlitzMax is a slower tool to develop in and the reason I dont use C++ is for development speed, with the key advantage of using Blitz gone i'm no longer interested in it.

I'm considering PureBasic for dual-platform/Macintosh programming, and staying with B3D on the PC for the time being until something better comes along.

I need to present to you that I found OOP way easier than the other style of basic programming. This MIGHT be because I spent a year coding in NWScript which is also OOP and is by far easy to use.

I know a lot of people here seem to repeat the same phrase that "BlitzMAX takes a long time to pickup" - but let me just be one of those who break that chain of rhetoric.

For what it's worth, I had a breakthrough in game development and I owe it to BltizMAX. Type programming is by far more comprehendable to my abstract brain.

If you have never used BlitzMAX and people are telling you its hard to pickup - don't listen to them. Pick it up yourself, and go through Wave's "make Asteroiuds in an hour or so" tutorial and / or my Vertical shooter tutorial and you'll realize that BMAX is EASY!!

Once you use just ONE or TWO tutorials you should understand OOP and then when BlitzMAX supports 3D you'll be happy you learned it.

Banshee,

Blitz3D is an easy language

Shaders are Not an easy language.

Why add such difficulties to an "easy-to-use" language?

HappySammy : Would you mind showing us your most advanced piece of B3D work to date? That way we can see that you've reached the limits of B3D and that you're not just another of the people who buy Blitz, do nothing with it and then decide it needs another 100 features before they can anything with it. It will give your opinions credence.

I have no problem with OOP - I welcome the object orientated approach - what I don't like is having to use a third party module for 3d, or write your own, which sort of defeats the purpose of buying a higher level games orientated language.

When/if BMax gets a native 3d module then you can talk about it being the successor to Blitz3d - currently its the successor to BlitzPlus.

Blitz3d was successful because it is stable, has very few bugs, is incredibly easy to learn , simple to use and has built in 3d commands. None of those could be used to describe BlitzMax at present.

BlitzMax, for me, was Stable, had very few bugs, I picked it up incredibly quick. It was easy to use. it even has 3D commands. Low level, but 3D non the less.

so strictly speaking, all but the B3D entity system is in Max.

Exactly - low level being the operative words.


Which is the cornerstone of Blitz3d simplicity and ease of use.


So did I - not everybody is in the same boat - some people still don't use types in Blitz3d !!

BlitzMAX is great. I find it very easy to use, and I picked it up quickly. I suppose there are many who refuse to do any 2D Graphics, I'm not one of them. I like a lot of 2D games out there, and I think they're easier to design. For one, you dont have to purchase models, and models can be expensive.

Since I'm not good at 3D models, and 2D Art is much cheaper than 3D Art, then I am going to make 2D Games. Since I find OOP easier about 1000x easier than arrays, then I'm in a win/win situation.

So blitzmax doesn't have 3D? Most people don't have uber machines that process the 3D games out there as it is. Considering 2D is still widely accepted and is successful, there's no need to complain!

3d games have been widely playable for well over 8 years.

I agree that 2d can be excellent though. Just look at Cletus Clay and Platypus! It's completely impossible to do that kind of work in 3d.


BlitzMax does do 3d. Many people have developed 3d engines for it and made wrappers for 3d engines. There is an open-source 3d engine called Trinity, (which I haven't heard anything about lately...) written entirely in OpenGL using native BlitzMax.


(Yay, First Post!)

Bmax is great just not geared towards 3D in the same way that B3D is. In time the 3D module for Bmax should have many of the benefits that B3D has at present.

And shaders need not be that hard, so long as Mark sticks to the established standards you should be able to use any number of WYSIWYG tools to make most of the things that your average artist can use to get the effects they like when doing pre rendered artwork.

Evak, the problem with shaders is OpenGL. It has a standard that isn't well supported by gfx card manufacturers. DX also has like 3 'standards', although one of those is supported well by BOTH companies, but it is still of course a dif language to the OpenGL standard. I think Mark may have to come up with another abstracted BlitzShader language/module.

How do you measure the 'level' of a language? Because most people I talk to measure it in abstraction level, that is the degree to which programming a language differs from programming the hardware. Thus OP-codes is a low-level language, because it's not avbstracted at all, Assembly is low-level, because it's still not very abstract, just easier to remember. Pascal and C are higher level (3rd generation) because they're structured languages that in many parts are abstracted from the underlying hardware (CPU and FPU). C++ is a higher level than Pascal, because it allows object orientation (which is an abstract concept that the CPU is not aware of). Java is even higher because it also provides abstraction for threads and automatical memory management.

Blitz3D is about Pascals level of abstraction, and BlitzMAX is about at Javas level of abstraction. Thus in conventional terms, BlitzMAX is a higher level language than Blitz3D.

Oh and by the way, bemoaning BlitzMAX because it doesn't "have a 3D module" is a non-issue. Like bemoaning your puppy because it isn't a dog yet. Sure we'd all love if everything we wanted was available when we want it, but hey, life sucks, get a @#!*ing helmet.

Not bemoaning, just don't use blitz products till the 3D module comes out or paid work demands it.

Abstraction has a more tangeable qualification: How many commands does it take too perform a complex task, B3D is the right "level" for my own personal tastes but it lacks some modern features.

BMax offers to make those features available to me - but works at a lower or "less abstract" level in that it requires more commands to do the same thing. It involves more typing, more thinking, and slower development times. On the plus side it can be optimised to a far greater standard which gives it greater flexibility.

Instead of doing things in a manner which B3D does quickly, you simply do things in a quick way and let BMax work around you - the downside is it's not what I want.

As for whether BMax is a good thing or not, well the bottom line is sales. I tried the BMax demo and felt it wasnt for me - i'm currently trying the PB demo instead. For regular 3D work on the PC i'm still using B3D and will until something comes along that lets me harness the new technologies. I have no immediate plans to buy BMax having tried the demo and finished a program in it.

That's not the same thing as wanting a 3D module that isn't out yet. I'm not entirely sure I understand and accept your abstraction argument that BMax is as simple as Java - it certainly a lot more complicated to learn than B3D, and yet young children can program in Java*.

*I've never personally been interested in Java, but it looks pretty crude and rudimentary when I view it.

I only hear bad things about game programming in Java. Allthough it's more often taught than C++ in many courses.

Well that's certainly not the case. Java is probably one of the most evolved mainstream programming languages around. The most being Python.

The only commercially successful independantly developed MMORPG (Puzzle Pirates) was written in Java.

There are some issues with writting games in Java, but most are (with Java 1.5 or better) regarding inexperienced programmers.